![]()
Collate, the Agentic Data Platform and the company behind OpenMetadata, today announced Ontology Studio, which lets enterprises document how their business works once, in an open-standard ontology they own, mapped to the data that implements it, and give that context to every AI agent that runs on that data to understand business meaning and increase accuracy. Ontology Studio is available in Collate 2.1.
Meaning Is the Moat: More Accurate Agents, Lower Costs
The knowledge of how a business runs is its moat in AI: which business concepts matter, how they relate, which rules apply, and which data implements them. An ontology writes that knowledge down in a form agents can read, and the context layer puts it in front of agents. Instead of guessing from table and column names, an agent can identify a Loan’s principal and term, find the right data, and apply the relevant business rules. That helps organizations get more accurate, consistent answers with less time and less token cost spent searching, retrying, and correcting mistakes. In a May 11, 2026 press release, Gartner®, Inc. predicted that “by 2027, organizations that prioritize semantics in AI-ready data will increase their agentic AI accuracy by up to 80% and reduce costs by up to 60%.”
Scout24, the DAX-listed company behind Germany’s leading real estate platform ImmoScout24, uses Collate as part of its data infrastructure to provide trusted business context for internal AI agents. By combining shared data definitions, AI-supported documentation and human verification, Scout24 is making business information more accessible and helping teams use AI to make better-informed decisions.
“AI agents are only as useful as the context behind their answers,” said Angelita Frozza Sanches, Head of Data at Scout24. “At Scout24, we’ve built a shared data foundation that connects business knowledge with our data. Our AI agents use that context to deliver more reliable answers, while confidence scores and feedback from our experts help us continuously improve. That’s what it takes to move AI from isolated experiments into everyday business use.”
Data and AI platform vendors are building ontology and context features into their own products that serve only the agents that run on that platform, and lock the enterprise’s knowledge into that vendor.
Writing that knowledge down in an open ontology, built on open standards and open source software, keeps the enterprise from being locked in. The company can inspect the ontology, move it, and use it with any platform or agent, so the moat stays with the company that built it.
“An ontology you can’t take with you belongs to your vendor,” said Jessica Talisman, Semantic Engineer, and Author of Ontology Pipeline. “The test is simple: can you inspect it, validate it, and export it in open standards? Collate Ontology Studio is built on OWL and SKOS and exports with its data mappings intact, so the organization keeps what it builds.”
How Collate Ontology Studio works
Collate Ontology Studio enables organizations to build, govern, and connect their own ontologies to data, all natively within Collate.
Model your business. Define concepts, typed properties, and named, directional relationships that reflect how your organization works. Native support for OWL and SKOS, the W3C standards for ontologies and controlled vocabularies, teams create their own concepts and relationship types, then review and publish versioned ontologies.
Connect concepts to data. Link business concepts to tables, columns, dashboards, and other data assets. Explore where each concept is represented in your data, with a clear distinction between technical lineage captured from data systems and business relationships defined in the ontology.
Keep your ontology open and portable. Import or export versioned ontologies in standard formats including OWL, Turtle, RDF/XML, and JSON-LD, preserving identifiers, definitions, and relationships for reuse in compatible tools and knowledge systems beyond Collate.
Give agents access. Through Collate’s Model Context Protocol (MCP) server, an agent can describe the ontology, traverse the relationships between concepts, and query it in SPARQL, the standard graph query language, on a read-only basis. These tools run on Collate’s knowledge graph projection, which is in beta and off by default.
Build context for agents quickly with industry starter packs. The Collate Ontology Library ships ready-to-use ontologies for FIBO in finance, HL7 FHIR in healthcare, GS1 and EPCIS in supply chain, ISA-95 in manufacturing, and HR Open Standards in HR, so a team can begin from the standard its industry already uses.
“Metadata, meaning and memory belong in one context platform,” said Sriharsha Chintalapani, Co-founder and CTO of Collate. “Ontology Studio keeps the ontology open and mapped to the data that implements it, so every agent a company runs works from the same definitions and business meaning that its people use.”
Availability
Ontology Studio will be generally available (GA) in Collate 2.1 later this month. Learn more in the Collate 2.1 Ontology Studio documentation.
About Collate
Collate is the Agentic Data Platform and the company behind OpenMetadata. Built on OpenMetadata’s open-source foundation and open standards, Collate brings your metadata, business meaning, and organizational memory together into trusted context so people and agents work from the same shared understanding, while your context stays portable and under your control. Collate automates data discovery, lineage, quality, observability, governance, analytics, and AI workflows, delivering answers with 75% fewer tokens. More than 4,000 organizations worldwide, including OpenAI, Yelp, and Experian, run their production data and AI workloads on Collate or OpenMetadata. Learn more at getcollate.io.
Source: Gartner® Press Release, “Gartner Says Lack of Semantics Causes Inaccurate AI Agents and Wasted Spending,” May 11, 2026, gartner.com/en/newsroom/press-releases/2026-05-11-gartner-says-lack-of-semantics-causes-inaccurate-artificial-intelligence-agents-and-wasted-spending. GARTNER is a registered trademark and service mark of Gartner, Inc. and/or its affiliates in the U.S. and internationally and is used herein with permission. All rights reserved.
View source version on businesswire.com: https://www.businesswire.com/news/home/20261117544775/en/
Media gallery